Backend Development 13 min read

Supporting Billions of Requests: The Evolution of Weibo Feed Cache Architecture

In this interview, Sina Weibo platform architect Chen Bo shares the history, design principles, and scaling techniques of Weibo's feed caching system, covering early LAMP setups, hybrid push‑pull models, service‑oriented migration, cloud integration, and practical advice for engineers.

Architecture Digest
Architecture Digest
Architecture Digest
Supporting Billions of Requests: The Evolution of Weibo Feed Cache Architecture

Editor’s note: Everyone’s growth path differs; some have notable products early, others learn through many pitfalls. This interview with Sina Weibo architect Chen Bo aims to inspire readers.

Event announcement: SDCC 2017 will be held in Shenzhen on June 10‑11, 2017. Chen Bo will speak on “Supporting Billions of Requests: The Evolution of Weibo Feed Cache Architecture.”

Guest introduction: Chen Bo is a platform architecture expert at Sina Weibo R&D. He joined Sina in 2008, worked on the IM backend, and since 2009 has been involved in the design and evolution of Weibo’s massive‑scale systems, focusing on infrastructure, middleware, and high‑availability architecture.

Career background: Chen has over ten years of development experience, initially working on Sina IM, then on the Weibo Feed platform, participating in all major versions and architectural improvements. He now focuses on feed infrastructure, cache middleware, and architecture optimization, with interests in service‑oriented architecture, distributed storage, hybrid cloud, big data, and high‑availability design.

Path to technology: Inspired by a university project that was acquired, Chen pursued self‑learning, certifications, and part‑time development, eventually joining research labs during graduate studies and moving into professional software development.

Work philosophy: Treat work as a career and an entrepreneurial venture; consider not only coding but also architecture evolution, industry trends, product innovation, and cost control (storage, power, cloud billing). Maintain curiosity and broaden technical horizons.

Reasons for staying at Sina: Open corporate culture, collaborative team environment with regular technical sharing, and the opportunity to work on impactful, large‑scale systems that drive social progress.

Definition of architecture and architect: Architecture is the overall structural design of a system, including abstract modules, interaction protocols, and design decisions, which must evolve with business needs. An architect designs the system’s blueprint, ensures high performance, availability, ease of implementation, operational simplicity, and extensibility.

Architect’s required abilities:

Strong coding skills and problem‑solving experience.

Excellent abstraction capability to translate business requirements into design blueprints.

Good communication and organization skills to discuss and disseminate architectural decisions.

Team collaboration and leadership to gain consensus and make timely technical choices.

Main responsibilities:

Translate business needs into architectural specifications, defining module interfaces, dependencies, performance, and error handling.

Organize discussions to ensure team understanding of the overall solution.

Assist project managers in planning and controlling development progress.

Determine foundational infrastructure and technologies, leading research and technical breakthroughs when needed.

Key architectural milestones of Weibo:

Early stage: LAMP stack, push‑based feed, MySQL storage, Memcached cache; limited stability and single‑point failures.

Mid‑stage: Migration from PHP to Java, modular Feed platform, hybrid push‑pull model, multi‑IDC deployment with unified degradation strategies and comprehensive monitoring.

Current stage: Service‑oriented and cloud‑native architecture; front‑end, business services, and resource layers are independently scalable; use of Motan RPC, ConfigService, CacheService, SLA/Trace/TouchStore for health monitoring; storage via MySQL, HBase, Redis, distributed file systems; greatly improved stability, security, and scalability.

Current architecture characteristics:

Good scalability: resources can be quickly expanded across private and public clouds.

High availability: multi‑active, no single point of failure; failures in one machine or IDC do not affect overall service.

Strong processing capacity: supports hundreds of billions of daily API calls and peak traffic spikes.

Comprehensive monitoring and alerting system for rapid issue detection and resolution.

Learning advice: Keep enthusiasm for new technologies, share and discuss knowledge, and apply what you learn to real projects.

Expectations for SDCC 2017: Insightful discussions on big‑data scenarios and practical solutions from industry leaders.

distributed systemsscalabilityBackend Developmentcache architectureWeibo
Architecture Digest
Written by

Architecture Digest

Focusing on Java backend development, covering application architecture from top-tier internet companies (high availability, high performance, high stability), big data, machine learning, Java architecture, and other popular fields.

0 followers
Reader feedback

How this landed with the community

login Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.